An image processing apparatus includes a full feature point detection section for detecting as a full feature point a point whose pixel value changes significantly along any line through the point, and a semi feature point detection section for detecting as a semi feature point a point whose pixel value hardly changes along one line through the point but changes significantly along other lines through the point.
Legal claims defining the scope of protection, as filed with the USPTO.
1. An image processing apparatus, comprising: a calculating section which calculates a gradient covariance matrix for each of a plurality of points in an image based on neighboring pixel values around the point, and which calculates multiple eigenvalues for each of the gradient covariance matrices; a full feature point detection section for detecting, and identifying as a full feature point, a point whose gradient covariance matrix has a minimum eigenvalue that is more than a certain amount; a semi feature point detection section for detecting, and identifying as a semi feature point, a point whose gradient covariance matrix has a largest maximum eigenvalue among points whose gradient covariance matrices have a ratio of a maximum eigenvalue to a minimum eigenvalue that is more than a predetermined amount; a first tracking section for tracking the identified full feature point between a first image and a second image; a second tracking section for tracking the identified semi feature point; and an alignment section for aligning the first and second images based on a tracking result of the first tracking section and the second tracking section.
2. The image processing apparatus according to claim 1 , wherein the alignment section comprises an evaluation section for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking section and the second tracking section, and the alignment section chooses the coordinate transform equation based on a judgment made by the evaluation section.
3. The image processing apparatus according to claim 2 , wherein the evaluation section weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
4. The image processing apparatus according to claim 1 , further comprising: an addition section for synthesizing one image from a plurality of images which have been subjected to alignment by the alignment section.
5. The image processing apparatus according to claim 4 , wherein the alignment section comprises an evaluation section for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking section and the second tracking section, and the alignment section chooses the coordinate transform equation based on a judgment made by the evaluation section.
6. The image processing apparatus according to claim 5 , wherein the evaluation section weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
7. An image processing method, comprising: a calculating step for calculating a gradient covariance matrix for each of a plurality of points in an image based on neighboring pixel values around the point, and for calculating multiple eigenvalues for each of the gradient covariance matrices; a full feature point detection step for detecting, and identifying as a full feature point, a point whose gradient covariance matrix has a minimum eigenvalue that is more than a certain amount; a semi feature point detection step for detecting, and identifying as a semi feature point, a point whose gradient covariance matrix has a largest maximum eigenvalue among points whose gradient covariance matrices have a ratio of a maximum eigenvalue to a minimum eigenvalue that is more than a predetermined amount; a first tracking step for tracking the identified full feature point between a first image and a second image; a second tracking step for tracking the identified semi feature point; and an alignment step for aligning the first and second images based on a tracking result of the first tracking step and the second tracking step.
8. The image processing method according to claim 7 , wherein the alignment step comprises an evaluation step for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking step and the second tracking step, and the alignment step comprises choosing the coordinate transform equation based on a judgment made by the evaluation step.
9. The image processing method according to claim 8 , wherein the evaluation step weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
10. The image processing method according to claim 7 , further comprising: an addition step for synthesizing one image from a plurality of images which have been subjected to alignment by the alignment step.
11. The image processing method according to claim 10 , wherein the alignment step comprises an evaluation step for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking steep and the second tracking step, and the alignment step comprises choosing the coordinate transform equation based on a judgment made by the evaluation step.
12. The image processing method according to claim 11 , wherein the evaluation step weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
13. A non-transitory computer-readable medium having an image processing program stored thereon that is executable by a computer to perform a process comprising: a calculating step for calculating a gradient covariance matrix for each of a plurality of points in an image based on neighboring pixel values around the point, and for calculating multiple eigenvalues for each of the gradient covariance matrices; a full feature point detection step for detecting, and identifying as a full feature point, a point whose gradient covariance matrix has a minimum eigenvalue that is more than a certain amount; a semi feature point detection step for detecting, and identifying as a semi feature point, a point whose gradient covariance matrix has a largest maximum eigenvalue among points whose gradient covariance matrices have a ratio of a maximum eigenvalue to a minimum eigenvalue that is more than a predetermined amount; a first tracking step for tracking the identified full feature point between a first image and a second image; a second tracking step for tracking the identified semi feature point; and an alignment step for aligning the first and second images based on a tracking result of the first tracking step and the second tracking step.
14. The computer-readable medium according to claim 13 , wherein the alignment step comprises an evaluation step for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking step and the second tracking step, and the alignment step comprises choosing the coordinate transform equation based on a judgment made by the evaluation step.
15. The computer-readable medium according to claim 14 , wherein the evaluation step weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
16. The computer-readable medium according to claim 13 , further comprising: an addition step for synthesizing one image from a plurality of images which have been subjected to alignment by the alignment step.
17. The computer-readable medium according to claim 16 , wherein the alignment step comprises an evaluation step for evaluating correctness of a calculated coordinate transform equation based on the tracking result of the first tracking step and the second tracking step, and the alignment step comprises choosing the coordinate transform equation based on a judgment made by the evaluation step.
18. The computer-readable medium according to claim 17 , wherein the evaluation step weights an evaluation related to the full feature point more than an evaluation related to the semi feature point.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 24, 2006
October 11, 2011
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.